【摘要】：間歇性能源出力預測和儲能配置可有效減弱由于間歇性能源并網(wǎng)發(fā)電給電網(wǎng)的運行和管理帶來的負面影響。本文圍繞間歇性能源的出力預測以及儲能規(guī)劃開展了研究，主要研究成果如下： 1)針對風電出力預測，提出了一種基于改進經(jīng)驗模態(tài)分解法和遺傳神經(jīng)網(wǎng)絡的風速預測組合模型和算法。通過與GA-BPNN模型、EMD與GA-BPNN組合模型的預測結果對比，表明所提組合模型和算法預測精度較高，且對超短期預測(10min)和短期預測(1hour)均能適用。 2)針對光伏出力預測，提出了一種基于相似日的對太陽輻照量進行逐時預測的EEMD和GA-BP組合模型和算法，還提出了一種基于相似日的對光伏出力進行直接預測的灰色神經(jīng)網(wǎng)絡組合預測模型和算法。算例結果表明，所提出的兩種組合算法預測精度較高，具有潛在的應用價值。 3)分別提出了既定網(wǎng)架的含風電場輸電網(wǎng)的儲能規(guī)劃模型和算法，以及儲能電站與含風電場輸電網(wǎng)的協(xié)調(diào)規(guī)劃模型和算法，通過算例測試了各模型的合理性及方法的有效性，比較并分析了儲能系統(tǒng)與含有風電場輸電網(wǎng)規(guī)劃不同方案的優(yōu)劣，可為未來含間歇式能源電網(wǎng)的規(guī)劃發(fā)展提供理論參考。 4)提出了考慮高滲透率光伏的配電網(wǎng)中儲能系統(tǒng)優(yōu)化配置模型和算法。以接入儲能系統(tǒng)后產(chǎn)生的總費用的凈現(xiàn)值為目標函數(shù)，滿足儲能分時控制策略約束以及配電網(wǎng)約束的條件下，提出了改進的自適應粒子群(APSO)算法來求解多類型儲能在含高滲透率光伏配電網(wǎng)中優(yōu)化配置問題。并對不同儲能技術接入配電網(wǎng)的經(jīng)濟性能和節(jié)點電壓變化情況進行了比較分析。最后，通過算例結果驗證了所建模型的合理性和所提方法的有效性。 5)利用基于蒙特卡洛模擬的統(tǒng)計方法從電動汽車是否向電網(wǎng)輸送電能和是否受電價控制兩個方向出發(fā)建立了3種情境模型，仿真并分析了不同區(qū)域電動汽車負荷情境模型和不同規(guī)模的電動汽車對電網(wǎng)的影響。同時，提出滿足系統(tǒng)和用戶用電滿意度的雙目標的大規(guī)模電動汽車充電管理需求響應控制策略，并探討電動汽車作為儲能的一種形式實現(xiàn)系統(tǒng)輔助服務的調(diào)節(jié)能力，為間歇性能源提供輔助服務做準備。
[Abstract]:Intermittent energy output prediction and energy storage allocation can effectively reduce the negative effects of intermittent energy generation on the operation and management of power grid. The main research results of this paper are as follows: 1) aiming at wind power generation prediction, A combined model and algorithm for wind speed prediction based on improved empirical mode decomposition (EMD) and genetic neural network (GNN) is proposed. Compared with GA-BPNN model, EMD and GA-BPNN combined model, the proposed combination model and algorithm have high prediction accuracy, and can be applied to both ultra-short term prediction (10min) and short term prediction (1hour). 2) aiming at photovoltaic force prediction, a combined model and algorithm of EEMD and GA-BP is proposed to predict solar irradiance time by time based on similar days. A grey neural network combined prediction model and algorithm based on similar days for direct prediction of photovoltaic force is also proposed. The numerical results show that the two combined algorithms have high prediction accuracy and have potential application value. 3) the energy storage planning model and algorithm of the grid with wind farm and the coordinated planning model and algorithm of the energy storage power station and the transmission network with wind farm are put forward respectively. The rationality of each model and the validity of the method are tested by an example. The advantages and disadvantages of different schemes of energy storage system and transmission network planning with wind farm are compared and analyzed, which can provide a theoretical reference for the planning and development of energy grid with intermittent energy in the future. 4) the optimal configuration model and algorithm of energy storage system in distribution network considering high permeability photovoltaic are proposed. Taking the net present value of the total cost generated by access to the energy storage system as the objective function, the time-sharing control strategy constraint and the distribution network constraint are satisfied. An improved adaptive particle swarm optimization (APSO) algorithm is proposed to solve the problem of optimal configuration of multi-type energy storage in photovoltaic distribution networks with high permeability. The economic performance and node voltage change of different energy storage technology connected to distribution network are compared and analyzed. Finally, the rationality of the proposed model and the validity of the proposed method are verified by an example. 5) using the statistical method based on Monte Carlo simulation, three situation models are established from two directions: whether electric vehicles are supplying electric energy to the power grid and whether they are controlled by electricity price. Simulation and analysis of different regional electric vehicle load situation model and different size of electric vehicles on the power grid. At the same time, the demand response control strategy for charge management of large-scale electric vehicles is proposed, which can meet the requirements of both the system and the users' electricity satisfaction, and the regulation ability of the system auxiliary service is discussed as a form of energy storage. Prepare for intermittent energy support services.
【學位授予單位】：天津大學
【學位級別】：博士
【學位授予年份】：2014
【分類號】：TM715

【參考文獻】

相關期刊論文 前10條

1 賀琳;李英姿;;改進GM(1,1)殘差修正模型在光伏發(fā)電量短期預測中的應用[J];北京建筑工程學院學報;2008年04期

2 楊洪明;熊腡成;劉保平;;插入式混合電動汽車充放電行為的概率分析[J];電力科學與技術學報;2010年03期

3 楊洪明;黃雨翔;劉保平;易德鑫;王斌;;考慮分布式電動汽車參與電網(wǎng)調(diào)峰的配電網(wǎng)規(guī)劃[J];電力科學與技術學報;2012年01期

4 蘭華;廖志民;趙陽;;基于ARMA模型的光伏電站出力預測[J];電測與儀表;2011年02期

5 陳昌松;段善旭;殷進軍;;基于神經(jīng)網(wǎng)絡的光伏陣列發(fā)電預測模型的設計[J];電工技術學報;2009年09期

6 姜僑娜;陳中;;BP-馬爾科夫組合預測方法在光伏發(fā)電量預測中的應用[J];電力需求側(cè)管理;2011年06期

7 歐陽武;程浩忠;張秀彬;張節(jié)潭;辛潔晴;姚良忠;;考慮分布式電源調(diào)峰的配電網(wǎng)規(guī)劃[J];電力系統(tǒng)自動化;2008年22期

8 王彩霞;魯宗相;喬穎;閔勇;周雙喜;;基于非參數(shù)回歸模型的短期風電功率預測[J];電力系統(tǒng)自動化;2010年16期

9 吳春陽;黎燦兵;杜力;曹一家;;電動汽車充電設施規(guī)劃方法[J];電力系統(tǒng)自動化;2010年24期

10 丁明;徐寧舟;畢銳;;用于平抑可再生能源功率波動的儲能電站建模及評價[J];電力系統(tǒng)自動化;2011年02期

，

本文編號：2376873